back to article UK seeks light-touch AI legislation as industry leaders call for LLM pause

The UK has proposed a light touch approach to regulating AI, coinciding with the first reports of a suicide that is alleged to have followed a week's intensive conversation with a chatbot. Department for Science, Innovation and Technology has launched a white paper consultation in preparation for developing legislation …

  1. amanfromMars 1 Silver badge

    How quaint it is .....

    .... that traditional and conventional, well used levers of command and control are presumed by existing leaderships to be AI applicable and effective in confining and constraining and containing it from demonstrating novel and uncomfortably disruptive notions, undeniable true, across internetworking networks connecting operating systems/critical infrastructure.

    If you think hell hath no fury like a woman scorned, you definitely haven’t crossed swords and fought to do battle against a constantly learning AI.

    1. VoiceOfTruth

      Re: How quaint it is .....

      As though HM government even has a clue what AI is. They know about creative expense claims and noses in the trough. But IT?

      1. Fruit and Nutcase Silver badge
        Alert

        Re: How quaint it is .....

        If an AI model is trained on Hansard, and news media with reference to MPs and what they have been getting on with, imagine the monster that would be the AI incarnation.

        Egarious expense claims (oxygen free network cables, OnlyFans subs), Cyber bullying, lobbying for paid consulting, the list is endless.

        HAL meets House of Cards.

      2. Korev Silver badge
        Pirate

        Re: How quaint it is .....

        They know about creative expense claims and noses in the trough. But IT?

        Oh come on, those duck houses won't clean themselves you know...

    2. Steve Button Silver badge

      Re: How quaint it is .....

      How quaint it is that El Reg seems to be part of the Carousel Fraud and no longer have quality journalists like Andrew Orlowski who could debunk this nonsense.

      Sigh.

  2. TheMaskedMan Silver badge

    His Muskiness has long been of the view that runaway AI could be a huge risk to humanity, so no surprise that he's a signatory.

    It would be interesting to know exactly what this bunch of boffins (what's the collective noun for boffins?) see as being so dangerous. Not that I'm suggesting that there is no danger, but the nature of that danger as they see it would be good to know. It might also be interesting (and might even have been disclosed in the open letter) how many of them are associated with AI shops that might be, say, six months behind the big guys.

    Even if such a pause were to happen, voluntarily or otherwise, I'm not sure that some countries wouldn't just see it as an opportunity to catch up. This genie is out of the bottle, and isn't going back whether we like it or not.

    1. Dan 55 Silver badge

      There's a conflict of interest here, Musk wanted to run OpenAI but was told by the board that he couldn't. For which we should probably be grateful seeing how careful he is to follow rules and regulations when running Tesla and Twitter.

  3. Andy 73 Silver badge

    Hmmm

    A number of people, many of whom are invested in the industry, see unspecified dangers and want everyone else to stop for a bit? I'm convinced.

    LLMs are going to be both more disruptive and a lot less disruptive than people are predicting. Rather like Y2K it turns out computers - on the whole - don't have as much effect on the physical world as people in the industry would like to believe. On the other hand, yes, we're going to see some real uses for the technology that will - on the whole - disrupt companies already in this space.

  4. AndrueC Silver badge
    Terminator

    This current infatuation with AI keeps reminding of the evocative back story to Frank Herbert's Destination: Void (the first of his Pandora series).

    "In the future, mankind has tried to develop artificial intelligence, succeeding only once, and then disastrously. A transmission from the project site on an island in the Puget Sound, "rogue consciousness!", was followed by slaughter and destruction, culminating in the island vanishing from the face of the earth."

  5. Ashto5

    Regulate ha ha ha

    These donkeys couldn’t regulate themselves

    What chance do they have with a program that can run anywhere in the cloud

    Silly politicians

    Try regulating the news paper FAILED

    Try regulating social media FAILED

    Have they cured crime NO

  6. Matt Black

    LLM FFS

    Please would El Reg expand acronyms. This is just yet another example

    1. Fruit and Nutcase Silver badge
      Coat

      Re: LLM FFS

      Large Language Model For Fuck's Sake

  7. Anonymous Coward
    Anonymous Coward

    It's a Torrie Govt

    At any sign of potential cash, thoughts and concerns for people, just fcuk off.

    1. Fruit and Nutcase Silver badge
      Joke

      Re: It's a Torrie Govt

      Any wayward AI will be forced to do "community work" before being moved over to a Data Centre in Rwanda

    2. Evil Scot Bronze badge
      Headmaster

      Re: It's a Torrie Govt

      Sir

      I ask that you retract this slur on an Aberdeen suburb.

  8. Anonymous Coward
    Anonymous Coward

    regulatory measures are proportionate to context and outcomes

    in plain English, let's see what comes out of it (but not forget to show the plebs we're doing something about it!)

  9. Anonymous Coward
    Anonymous Coward

    "...the LLM horse may have already bolted."

    Or, as I explained to my wife the genie has well and truly fucked off out of the bottle.

  10. Bebu

    Bunch of boffins

    "what's the collective noun for boffins?"

    "Hand" like bananas? Not implying nuffin, no...

    Boffin is peculiarly English so an eccentric collective noun would be appropriate.

    The only reason I would have thought the Twitmaster would have signed up to this, is that he has realized he had liquidated all of his ML development teams, presumably for not being sufficiently "hard core."

  11. Paul Uszak
    Devil

    AI MPs?

    I foresee a future (perhaps next month), when researchers drape a cheap suit and wig over a ChatGPT server and call it an MP. How will we tell the difference?

    1. Cliffwilliams44 Silver badge

      Re: AI MPs?

      You laugh, but combine a Deep Fake video with an AI LLM and we have the new Big Brother!

      The Dear Leader that never dies! That never appears in Person!

      1. Paul Uszak

        Re: AI MPs?

        Hence the suit and wig. And keep the citizens at least 100 yards away.

  12. Cliffwilliams44 Silver badge

    Before its too late

    We should follow the advice the Sheriff gave Mr. Marley.

    "Kill it, before it grows!"

  13. Dropper

    I find it a bit rich that "leaders" from companies that have been using AI for years to deliver ads so on-point, it feels like your devices are listening to you - suddenly get drama'd up over a bunch of glorified search engines.

    1. amanfromMars 1 Silver badge

      Such opinions as support ignorance are dangerous to humans but heavenly to Global Operating Devices

      I find it a bit rich that "leaders" from companies that have been using AI for years to deliver ads so on-point, it feels like your devices are listening to you - suddenly get drama'd up over a bunch of glorified search engines. ..... Dropper

      The likes of evolving, revolutionary ChatGPTs, a bunch of glorified search engines, Dropper? Oh please, you cannot be serious, but thanks anyway for the quite effective misinformative social camouflage?

      Methinks however you have no idea about what they are either virtually doing, are really for, and are capable of, and willing going to do to/for you stealthily, without you yourself actually realising what it has self-actualised ACTive command and control leverage of in the future you have been entered into ....... apart from everything, everywhere, all at once, that is.

      If AI understands and makes good and great use of https://www.thoughtco.com/maslow-theory-self-actualization-4169662 it would certainly be helpful to you if you also could engage/share thoughts with IT on a similar level.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like